Re: Interesting JiffyDOS / C128 problem

From: Francesco Messineo <francesco.messineo_at_gmail.com>
Date: Tue, 11 Aug 2020 14:57:09 +0200
Message-ID: <CAESs-_xsQmf--e3eNKhxtAaRgBDoURVk7A+CifFWUNx=Jxo-8w_at_mail.gmail.com>
Hi Jeff,
a couple of ideas.
1) Are these "ROMs" actually EPROMs? I've found quite a few EPROMs
programmed with these modern USB programmers not to have valid TTL
thresholds when read on a proper programmer (or when used on a old TTL
environment). When programmed marginally, such an EPROM may or may not
work on a given machine. However I have also to mention that the bad
parts that I've seen always were the high voltage and low density ones
(2732, 2532, 2716 etc...).
2) There could be some "data hold" timing issue where a fast ROM goes
too soon into data bus threestate after the enable has been lifted.
That might depend also on data bus transceiver's timings if present.
I would exclude a CIA problem otherwise all the different ROMs would
behave in the same manner.
I once found an 8500 CPU that would hang on some very specific chunks
of code, when it was warm. Moving that CPU from a PAL machine to an
NTSC machine made the hang disappear for as much as I could test. So
strange faults do indeed exist, but I would first suspect some logic
theshold or timing problem.

Frank

On Tue, Aug 11, 2020 at 2:44 PM Jeffrey Birt <birt_j_at_soigeneris.com> wrote:
>
> Hi everyone,
>
>
>
> I had someone email me an interesting question about a C128 with JiffyDOS. He has two flat C128s, both working. He put JiffyDOS in one and it works fine. He put JiffyDOS in the 2nd machine and got a black screen. He put the original ROM back in the 2nd one and it booted fine again. He bought another JiffyDOS ROM, installed it in 2nd machine and gets a garbage screen. He put the original ROM back in and it works fine. So, it does seem that his second C128 machine does not play nice with JiffyDOS.
>
>
>
> The reason he emailed me was he saw the video I did on the VIC tower mod and he wondered if his second C128 had that mod board and it was causing the problem. I told him that the VIC tower mod board only had an effect on RAM write timing, so it more than likely was not an issue in this case. It did make me wonder what the problem might be though.
>
>
>
> My thoughts were that on the C128 CIA#2 is what talks to everything on the IEC bus. The three IEC bus signals that come out of the CIA#2 also go though some inverter/buffers. I’m just guessing but wonder if there is something not quite right with CIA#2. A quick test would be to swap CIA#1 and CIA#2 and see if the problem changes. CIA#1 deals mostly with the keyboard, joysticks, etc.
>
>
>
> I have done some web searching and did not find a problem similar to this mentioned. It makes me curious what the problem might be so I thought I would mention it here to see if anyone had any ideas that might help him out.
>
>
>
> Thanks,
>
> Jeff Birt
>
>
>
>
>
>
>
>
>
>
>
>
Received on 2020-08-11 15:01:08

Archive generated by hypermail 2.3.0.